Recent expression profile analyses revealed that lung adenocarcinomas can be divided into subgroups with diverse pathological features. Because cellular heterogeneity of tumors can confound these analyses, we used laser capture microdissection and microarray expression analysis to characterize the molecular profiles of lung adenocarcinomas. We found 45 genes delineating smokers and nonsmokers that were located at chromosomal loci frequently altered in non-small cell lung cancers including 3p21.3, and 27 genes which were differentially expressed between survivors and nonsurvivors 5 years after surgery. These results are consistent with the hypothesis that the abnormal expression of genes involved in maintaining the mitotic spindle checkpoint and genomic stability, e.g., hBUB3, hZW10, and APC2, contribute to the molecular pathogenesis and tumor progression of tobacco smoke-induced adenocarcinoma of the lung. We are currently investigating the functional and genomic consequences of dysregulation of hBUB3, hZW10 and APC2. We also are molecular profiling lung tumors with neuroendocrine differentiation including typical and atypical carcinoids, small cell carcinomas, and a subset of adenocarcinoma. Poor survival gene expression sets have been identified and internally validated by RT-PCR and protein analyses and cross-validated with an additional tumor cohort. Free radical-induced cellular stress contributes to cancer during chronic inflammation. Because the p53 pathway can be activated by free radicals, we have investigated mechanisms of p53 activation by the free radical, NO?. NO? from donor drugs induced both ataxia-telangiectasia mutated (ATM) and ATR-dependent p53 posttranslational modifications, leading to an increase in p53 dependent transcriptional targets and G2/M cell cycle checkpoint. Such modifications were also identified in cells cocultured with NO?-releasing macrophages. In noncancerous colon tissues from patients with ulcerative colitis (a cancer-prone chronic inflammatory disease), inducible NO? synthase protein levels were positively correlated with p53 serine 15 phosphorylation levels. Immunostaining of HDM-1 and p21WAF1 was consistent with transcriptionally active p53. Our study highlights a pivotal role of NO? in the induction of cellular stress and the activation of a p53 response pathway during chronic inflammation. Ulcerative colitis (UC) is associated with chromosomal and microsatellite instability. To investigate the mechanism of this instability, we examined paired tissues from noncancerous colons of 30 UC patients to determine the ativity of two base excision repair (BER) enzymes, AAG, the major 3-methyladenine DNA glycosylase (3-MeA) DNA glycosylase and APE1, the major apurinic site (AP) endonuclease, and to determine the prevalence of microsatellite instability (MSI). AAG and APE1 were significantly increased in the UC colon epithelium undergoing elevated inflammation and MSI was positively correlated with their enzymatic activities. These latter results are supported by mechanistic studies with Leona Samson and coworkers using yeast and human cell models, in which, overexpression of the AAG glycosylase and/or AP endonuclease generated frameshift mutations and MSI. Furthermore, we have shown with Varda Rotter and coworkers that increased NO? enhances AAG glycosylase activity in an animal model. Taken together, our results are consistent with the hypothesis that the adaptive-imbalanced increase in AAG and APE1 is a novel mechanism contributing to MSI in patients with UC, and may extend to other diseases of chronic inflammation or those with MSI of unknown etiology. Animal models are an integral facet of a comprehensive strategy to investigate human carcinogenesis including the role of inflammation and oxidative stress in carcinogenesis. Therefore, we are investigating animal models with single, double, or triple gene knockouts of p53, NOS2, and cyclooxygenase 1 or 2, to conduct translational studies of gene-microenvironment in carcinogenesis. Our initial results indicate that basal levels of NOS2 are protective against spontaneous lymphoma and sarcoma formation in p53 homozygous and heterozygous knockout mice and one mechanism of this protection is the role of NO? in apoptotic and tumor immunity pathways. Cyclooxygenase-2 (COX-2) is overexpressed in multiple types of human cancer and is considered a molecular target for both cancer preventive and therapeutic agents. We have also discovered that COX2 is regulated by both the WNT and RAS pathways that function in cooperative transcriptional and posttranscriptional mechanisms. Mutations in the adenomatous polyposis coli (APC) gene and K-ras occur in the majority of human colorectal cancers. Loss of functional APC protein activates the WNT signal transduction pathway, allowing the nuclear accumulation of ?-catenin, which then binds to T-cell factor-4 (Tcf-4), causing increased transcriptional activation of downstream target genes. We have investigated the hypothesis that the activation of the WNT pathway regulates COX-2. COX-2 was downregulated after the induction of full-length APC in the HT29-APC cell line. We identified a Tcf-4-binding element (TBE) in the COX-2 promoter that specifically bound to Tcf-4 in an electrophoretic mobility shift assay. COX-2 promoter luciferase activity is downregulated by APC in a promoter-reporter construct containing TBE, but not with mutant TBE. Mutant ?-catenin expression upregulated the COX-2 promoter activity and the endogenous COX-2 mRNA expression in HuH7 cells, hepatocellular carcinoma cell line, which is partially abrogated by cotransfection with a dominant-negative Tcf-4 expression vector. Although ?-catenin alone did not increase COX-2 protein to detectable levels in HuH7 cells, coexpression of both mutant ?-catenin and mutant K-ras increased COX-2 protein expression, which is consistent with the previous reports that K-ras can stabilize COX-2 mRNA. Taken together, our data support the hypothesis that COX-2 is downregulated by APC and upregulated by nuclear ?-catenin accumulation, and additionally implicate the WNT signal transduction pathway in colon and liver carcinogenesis. These data further identify the complex interaction of the NO?, WNT, RAS, and COX2 pathways.